CODEWITHSUNDEEP
sas
Milad
Milad

Reputation: 339

Overwriting the optimal values of parameters after running PROC NLP in SAS

How can I get the following code to overwrite the optimal values of my parameters (which is mean and sigma) in the parameter dataset?

data parameter(type=est); 
        Keep _type_ sigma mean;
        _type_='parms'; sigma=0.1; mean=0.1;

data data; 
        input x @@; 
        datalines;
        1 3 4 5 7;


proc nlp data=Data NOMISS tech=tr inest=parameter Pcov phes;
    max loglik;
    parms sigma mean;
    loglik=-0.5*((x-mean)/sigma)**2-log(sigma);
run;

Upvotes: 0

Views: 93

Answers (1)

Rick
Rick

Reputation: 1210

On the PROC NLIN statement, use the OUTEST= option to create an output data set that contains the optimal parameter values, among other statistics.

Although I don't recommend directly overwriting the INEST= data set, you can do it by using the same name on the OUTEST= statement. Use a WHERE clause and a KEEP clause to make the output data set look exactly like the input data. For example:

proc nlp data=D NOMISS tech=tr inest=parameter Pcov phes 
     outest=ParamOpt(where=(_TYPE_="PARMS") keep=_type_ sigma mean);

I've used 'ParamOpt' as the name of the output data set, but you can use 'Parameter' to overwrite the input data set.

Upvotes: 1

Related Questions